Newest papers

Adaptive Skill Granularity for Coding Agents: Runtime Routing and Macro-to-Micro Recovery

Dingxuan Zhang, Yanhui Jiang, Jun Wang, Hefeng Zhou · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.31461 · cited by 0 works (OpenAlex) · open-access PDF

Spanish

Las habilidades reutilizables permiten a los agentes del modelo de lenguaje grande empaquetar procedimientos específicos de tareas, affordances de herramientas y orientación de ejecución en unidades modulares. El trabajo anterior ha estudiado en gran medida cómo los agentes recuperan, componen o sintetizan habilidades antes de la ejecución. En los agentes de codificación, sin embargo, la ejecución plantea un problema de control de ejecución-abstracción adicional: la misma tarea puede ser resuelta a través de llamadas de herramientas primitivas, habilidades macro o ejecución híbrida que comienza con una macro y regresa a la recuperación de estado fijo. Introducimos la granularidad de habilidades adaptativas, un marco que trata la granularidad de habilidades como una política de acción-abstracción condicionada a la estructura

Machine translation (Meta M2M100); the English original follows.

Abstract

Reusable skills allow large language model agents to package task-specific procedures, tool affordances, and execution guidance into modular units. Prior work has largely studied how agents retrieve, compose, or synthesize skills before execution. In coding agents, however, execution poses an additional runtime control problem: the same task may be solved through primitive tool calls, macro skills, or hybrid execution that starts with a macro and falls back to micro-level recovery. We introduce Adaptive Skill Granularity, a framework that treats skill granularity as an action-abstraction policy conditioned on task structure, execution state, and macro-skill failures. The system combines a feature-based granularity controller and a finite-state recovery executor. We evaluate the framework with a deterministic coding-agent protocol using verifier-defined success, total cost, and process metrics such as macro adoption, failure recovery, route match, and transition correctness. Results show that fixed macro execution is efficient when applicable but brittle under mismatch, while runtime routing and explicit macro-to-micro recovery improve robustness and recover controlled macro failures that fixed-granularity execution leaves unresolved. Our findings suggest that skill libraries for coding agents should be treated not only as retrievable resources, but also as controllable, verifiable, and recoverable execution abstractions.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Granularity, Modular design, Macro, Coding (social sciences), Distributed computing

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange